From 9f959fc6ed50cc2e6398dfd2ac8b066638ba3e56 Mon Sep 17 00:00:00 2001
From: Rodolphe Marques <rodolphe@ascribe.io>
Date: Mon, 9 May 2016 16:45:09 +0200
Subject: [PATCH 1/3] Add command to configure number of shards.

Changed aws deployment script to automatically set the number of shards.
Created tests
---
 bigchaindb/commands/bigchain.py | 14 ++++++++++++++
 deploy-cluster-aws/awsdeploy.sh |  1 +
 deploy-cluster-aws/fabfile.py   |  6 ++++++
 docs/source/bigchaindb-cli.md   |  7 +++++++
 tests/test_commands.py          | 22 +++++++++++++++++++++-
 5 files changed, 49 insertions(+), 1 deletion(-)

diff --git a/bigchaindb/commands/bigchain.py b/bigchaindb/commands/bigchain.py
index b37a3812..e6b4314d 100644
--- a/bigchaindb/commands/bigchain.py
+++ b/bigchaindb/commands/bigchain.py
@@ -13,6 +13,7 @@ import builtins
 
 import logstats
 
+import rethinkdb as r
 
 import bigchaindb
 import bigchaindb.config_utils
@@ -203,6 +204,12 @@ def run_load(args):
     workers.start()
 
 
+def run_sharding(args):
+    b = bigchaindb.Bigchain()
+    r.table('bigchain').reconfigure(shards=args.num_shards, replicas=1).run(b.conn)
+    r.table('backlog').reconfigure(shards=args.num_shards, replicas=1).run(b.conn)
+
+
 def main():
     parser = argparse.ArgumentParser(
         description='Control your BigchainDB node.',
@@ -243,6 +250,13 @@ def main():
     subparsers.add_parser('start',
                           help='Start BigchainDB')
 
+    # parser for configuring the number of shards
+    sharding_parser = subparsers.add_parser('sharding',
+                                            help='Configure number of shards')
+
+    sharding_parser.add_argument('num_shards', metavar='num_shards', type=int, default=1,
+                                 help='Number of shards')
+
     load_parser = subparsers.add_parser('load',
                                         help='Write transactions to the backlog')
 
diff --git a/deploy-cluster-aws/awsdeploy.sh b/deploy-cluster-aws/awsdeploy.sh
index 3ce621d9..b68b0a5e 100755
--- a/deploy-cluster-aws/awsdeploy.sh
+++ b/deploy-cluster-aws/awsdeploy.sh
@@ -131,6 +131,7 @@ if [ "$WHAT_TO_DEPLOY" == "servers" ]; then
     # this will only be sent to one of the nodes, see the
     # definition of init_bigchaindb() in fabfile.py to see why.
     fab init_bigchaindb
+    fab configure_sharding:$NUM_NODES
 else
     # Deploying clients
     # The only thing to configure on clients is the api_endpoint
diff --git a/deploy-cluster-aws/fabfile.py b/deploy-cluster-aws/fabfile.py
index 1d481111..581a1c1b 100644
--- a/deploy-cluster-aws/fabfile.py
+++ b/deploy-cluster-aws/fabfile.py
@@ -166,6 +166,12 @@ def init_bigchaindb():
     run('bigchaindb init', pty=False)
 
 
+@task
+@hosts(public_dns_names[0])
+def configure_sharding(num_shards):
+    run('bigchaindb sharding {}'.format(num_shards))
+
+
 # Start BigchainDB using screen
 @task
 @parallel
diff --git a/docs/source/bigchaindb-cli.md b/docs/source/bigchaindb-cli.md
index 229d5b4c..43f8dbb0 100644
--- a/docs/source/bigchaindb-cli.md
+++ b/docs/source/bigchaindb-cli.md
@@ -43,3 +43,10 @@ This command is used to run benchmarking tests. You can learn more about it usin
 ```text
 $ bigchaindb load -h
 ```
+
+### bigchaindb sharding
+
+This command is used to configure the number of shards in the underlying datastore, for example:
+```text
+$ bigchaindb sharding 3
+```
\ No newline at end of file
diff --git a/tests/test_commands.py b/tests/test_commands.py
index 99a4f466..f90bc503 100644
--- a/tests/test_commands.py
+++ b/tests/test_commands.py
@@ -1,11 +1,12 @@
 import json
 from unittest.mock import Mock, patch
 from argparse import Namespace
-from pprint import pprint
 import copy
 
 import pytest
 
+from tests.db.conftest import setup_database
+
 
 @pytest.fixture
 def mock_run_configure(monkeypatch):
@@ -225,3 +226,22 @@ def test_start_rethinkdb_exits_when_cannot_start(mock_popen):
     with pytest.raises(exceptions.StartupError):
         utils.start_rethinkdb()
 
+
+def test_configure_sharding(b):
+    import rethinkdb as r
+    from bigchaindb.commands.bigchain import run_sharding
+
+    # change number of shards
+    args = Namespace(num_shards=3)
+    run_sharding(args)
+
+    # retrieve table configuration
+    table_config = list(r.db('rethinkdb')
+                        .table('table_config')
+                        .filter(r.row['db'] == b.dbname)
+                        .run(b.conn))
+
+    # check shard configuration
+    for table in table_config:
+        if table['name'] in ['backlog', 'bigchain']:
+            assert len(table['shards']) == 3

From f09dacdd118596cf3cbe36394a647e66459a81c0 Mon Sep 17 00:00:00 2001
From: Rodolphe Marques <rodolphe@ascribe.io>
Date: Mon, 9 May 2016 17:31:02 +0200
Subject: [PATCH 2/3] renamed sharding to set-shards

---
 bigchaindb/commands/bigchain.py | 4 ++--
 tests/test_commands.py          | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/bigchaindb/commands/bigchain.py b/bigchaindb/commands/bigchain.py
index e6b4314d..28831642 100644
--- a/bigchaindb/commands/bigchain.py
+++ b/bigchaindb/commands/bigchain.py
@@ -204,7 +204,7 @@ def run_load(args):
     workers.start()
 
 
-def run_sharding(args):
+def run_set_shards(args):
     b = bigchaindb.Bigchain()
     r.table('bigchain').reconfigure(shards=args.num_shards, replicas=1).run(b.conn)
     r.table('backlog').reconfigure(shards=args.num_shards, replicas=1).run(b.conn)
@@ -251,7 +251,7 @@ def main():
                           help='Start BigchainDB')
 
     # parser for configuring the number of shards
-    sharding_parser = subparsers.add_parser('sharding',
+    sharding_parser = subparsers.add_parser('set-shards',
                                             help='Configure number of shards')
 
     sharding_parser.add_argument('num_shards', metavar='num_shards', type=int, default=1,
diff --git a/tests/test_commands.py b/tests/test_commands.py
index f90bc503..4b35edbb 100644
--- a/tests/test_commands.py
+++ b/tests/test_commands.py
@@ -229,11 +229,11 @@ def test_start_rethinkdb_exits_when_cannot_start(mock_popen):
 
 def test_configure_sharding(b):
     import rethinkdb as r
-    from bigchaindb.commands.bigchain import run_sharding
+    from bigchaindb.commands.bigchain import run_set_shards
 
     # change number of shards
     args = Namespace(num_shards=3)
-    run_sharding(args)
+    run_set_shards(args)
 
     # retrieve table configuration
     table_config = list(r.db('rethinkdb')

From e21282818262974a9224183d9ecf607452467736 Mon Sep 17 00:00:00 2001
From: troymc <troy@ascribe.io>
Date: Mon, 9 May 2016 18:13:20 +0200
Subject: [PATCH 3/3] Updated awsdeploy, fabfile, docs, test for bigchaindb
 set-shards

---
 deploy-cluster-aws/awsdeploy.sh | 2 +-
 deploy-cluster-aws/fabfile.py   | 5 +++--
 docs/source/bigchaindb-cli.md   | 6 +++---
 tests/test_commands.py          | 6 +++---
 4 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/deploy-cluster-aws/awsdeploy.sh b/deploy-cluster-aws/awsdeploy.sh
index b68b0a5e..6290d01e 100755
--- a/deploy-cluster-aws/awsdeploy.sh
+++ b/deploy-cluster-aws/awsdeploy.sh
@@ -131,7 +131,7 @@ if [ "$WHAT_TO_DEPLOY" == "servers" ]; then
     # this will only be sent to one of the nodes, see the
     # definition of init_bigchaindb() in fabfile.py to see why.
     fab init_bigchaindb
-    fab configure_sharding:$NUM_NODES
+    fab set_shards:$NUM_NODES
 else
     # Deploying clients
     # The only thing to configure on clients is the api_endpoint
diff --git a/deploy-cluster-aws/fabfile.py b/deploy-cluster-aws/fabfile.py
index 581a1c1b..9e4a1d47 100644
--- a/deploy-cluster-aws/fabfile.py
+++ b/deploy-cluster-aws/fabfile.py
@@ -166,10 +166,11 @@ def init_bigchaindb():
     run('bigchaindb init', pty=False)
 
 
+# Set the number of shards (in the backlog and bigchain tables)
 @task
 @hosts(public_dns_names[0])
-def configure_sharding(num_shards):
-    run('bigchaindb sharding {}'.format(num_shards))
+def set_shards(num_shards):
+    run('bigchaindb set-shards {}'.format(num_shards))
 
 
 # Start BigchainDB using screen
diff --git a/docs/source/bigchaindb-cli.md b/docs/source/bigchaindb-cli.md
index 43f8dbb0..ad9fd5a0 100644
--- a/docs/source/bigchaindb-cli.md
+++ b/docs/source/bigchaindb-cli.md
@@ -44,9 +44,9 @@ This command is used to run benchmarking tests. You can learn more about it usin
 $ bigchaindb load -h
 ```
 
-### bigchaindb sharding
+### bigchaindb set-shards
 
-This command is used to configure the number of shards in the underlying datastore, for example:
+This command is used to set the number of shards in the underlying datastore. For example, the following command will set the number of shards to four:
 ```text
-$ bigchaindb sharding 3
+$ bigchaindb set-shards 4
 ```
\ No newline at end of file
diff --git a/tests/test_commands.py b/tests/test_commands.py
index 4b35edbb..12c1350e 100644
--- a/tests/test_commands.py
+++ b/tests/test_commands.py
@@ -227,11 +227,11 @@ def test_start_rethinkdb_exits_when_cannot_start(mock_popen):
         utils.start_rethinkdb()
 
 
-def test_configure_sharding(b):
+def test_set_shards(b):
     import rethinkdb as r
     from bigchaindb.commands.bigchain import run_set_shards
 
-    # change number of shards
+    # set the number of shards
     args = Namespace(num_shards=3)
     run_set_shards(args)
 
@@ -241,7 +241,7 @@ def test_configure_sharding(b):
                         .filter(r.row['db'] == b.dbname)
                         .run(b.conn))
 
-    # check shard configuration
+    # check that the number of shards got set to the correct value
     for table in table_config:
         if table['name'] in ['backlog', 'bigchain']:
             assert len(table['shards']) == 3